Diabetes and Cirrhosis: What Changes When the Liver Is Scarred
Once liver disease reaches the scarring stage, three ordinary parts of diabetes care stop working the way they used to — the HbA1c test, the assumption that highs are the main danger, and the medicine list.

Fatty liver is the liver problem most people with type 2 diabetes hear about, and at that stage the advice runs in a familiar direction — lose some weight, cut liquid sugar, walk. This article is about what happens further along the road, once a liver has actually scarred: cirrhosis, from any cause. Hepatitis C, hepatitis B, years of fatty liver, alcohol — the cause matters for treating the liver, but for blood sugar the scarring itself is what changes things.
And it changes them more than most people expect. Diabetes and cirrhosis are not two conditions running side by side; the second one quietly rewrites how you manage the first. Three things in particular stop behaving the way they used to.
1. Your HbA1c may be reading low
HbA1c works by measuring how much sugar has stuck to your red blood cells over their roughly three-month lifespan. Cirrhosis shortens that lifespan — an enlarged spleen destroys red cells faster, and anaemia is common — so the cells are replaced before they have finished collecting the sugar. The number that comes back is lower than your actual blood sugar warrants.
This is measurable rather than theoretical. In a study published in Digestive Diseases and Sciences (opens in a new tab), HbA1c tracked continuous glucose monitoring data closely in people with diabetes alone (R² 0.94) but poorly in people who had diabetes and cirrhosis together (R² 0.52) — a reassuring HbA1c in that second group simply carried much less information about what glucose was actually doing.
If you have cirrhosis, treat your HbA1c as one weak signal rather than the verdict. Your own before- and after-meal readings, an oral glucose tolerance test, or a CGM sensor if you can access one will show what the lab number hides. A fructosamine test, which reflects the past two to four weeks, is sometimes used as an alternative — though it did not perform better than HbA1c in the study above, so it is a conversation with your doctor, not a straight swap.
The same shortened-red-cell problem is why diabetes caused by a scarred liver is often missed at diagnosis. Fasting glucose in this situation is frequently normal; the abnormality shows up after eating.
2. Lows become the bigger risk
This is the reversal that catches people out. In ordinary type 2 diabetes, the day-to-day worry is highs. With cirrhosis, hypoglycaemia moves up the list — because the liver is the organ that rescues you from a low.
Two jobs it normally does: it holds a store of glycogen it can release when your sugar drops, and it manufactures fresh glucose when that store runs out. A scarred liver does both badly, and it also clears insulin from the blood less efficiently, so insulin lingers. Add the muscle loss and poor appetite that often come with advanced liver disease and there is very little reserve to fall back on.
So a dose of medicine that was safe for years may not be safe now, and a missed meal costs more than it used to. If you take insulin or a sulfonylurea, this is worth raising with your doctor as soon as cirrhosis is diagnosed rather than after the first bad low.
3. Long gaps without food stop being harmless
A healthy body treats an overnight fast as unremarkable. A cirrhotic body does not — the EASL guidelines on nutrition in chronic liver disease (opens in a new tab) describe accelerated starvation, where the body starts breaking down protein for fuel much sooner than it should. Their recommended pattern is three meals plus three snacks, with the late-evening snack singled out as the most important one because it covers the longest gap of the day, between dinner and breakfast.
For someone managing diabetes that sounds contradictory — eating at bedtime is usually the habit people are trying to break. The guidance for people with both conditions is a low-sugar late-evening snack, not none. Protein is the other surprise: the recommendation is 1.2 to 1.5 g per kg of body weight per day, well above a normal diet, because muscle loss is one of the things that shortens life in cirrhosis. Older advice to restrict protein in liver disease has been reversed for most people.
Voluntary fasting deserves a direct mention here. Ramadan fasting with diabetes already needs a risk assessment; cirrhosis on top of it changes that assessment substantially, and it is a decision to make with your doctor well before Ramadan begins, not in the first week.
What happens to your medicines
The instinct — and sometimes the reflex of a busy clinic — is to stop everything when liver tests look abnormal. That is often the wrong move, metformin in particular. A review of hepatogenous diabetes treatment (opens in a new tab) reports that a meta-analysis found no significant association between metformin and lactic acidosis in this group, and that people who stayed on metformin had a longer median survival than those who did not, alongside a 7% yearly reduction in the risk of primary liver cancer.
Where the medicine list does change is with how much liver function remains. Cirrhosis is graded — Child-Pugh A, B or C — and that grade, not the diagnosis alone, drives most of these decisions.
| Medicine | How cirrhosis changes the picture |
|---|---|
| Metformin | Remains the backbone of treatment in compensated cirrhosis. Usually stopped only in decompensated disease or kidney impairment — not on abnormal liver enzymes alone. |
| Sulfonylureas (gliclazide, glimepiride) | Higher hypoglycaemia risk: the liver clears them more slowly, and low albumin leaves more free drug active. Caution or avoidance as liver reserve declines. |
| Insulin | The mainstay in decompensated cirrhosis, but with conservative doses and frequent adjustment, because insulin is cleared more slowly and lows are harder to recover from. |
| GLP-1 medicines (Ozempic, Mounjaro) and SGLT2 inhibitors | Promising in compensated disease, especially alongside fatty liver — but GLP-1 use is currently limited to Child-Pugh A, and fluid-retaining drugs need caution. |
| Pioglitazone | Used at reduced dose in chronic liver disease, with a ceiling of 30 mg/day. |
| Alpha-glucosidase inhibitors (acarbose) | Barely absorbed and almost not metabolised by the liver, which makes them one of the better-tolerated options here. |
This table is for recognising which questions are worth asking, not for adjusting anything. Every one of these decisions depends on your Child-Pugh grade, your kidney function and your current readings — all things your doctor has and this article does not.
Why this matters more in Pakistan
Cirrhosis is common here in a way it is not in many countries, largely because hepatitis C is. Diabetes appears in about 30.7% of people with cirrhosis overall, and in 61.2% of those with the most advanced disease, according to the pooled figures in a review of hepatogenous diabetes (opens in a new tab). The practical consequence is that a great many people in Pakistan are managing both conditions — often without anyone having explained that the second one changes the first.
If you have diabetes and have never been tested for hepatitis B and C, that is the single most useful test to ask for, because it is the difference between knowing your liver's situation and guessing at it.
When HbA1c is unreliable, your own log becomes the primary record — and the pattern to watch for is a normal fasting number with high readings after meals, or lows appearing where they never used to. Logging in Diatic keeps that pattern visible, so you and your doctor are looking at real data rather than one questionable lab value.
Frequently asked questions
My doctor stopped my metformin because my liver enzymes were high. Was that wrong?
Not necessarily — there may have been a reason specific to you, such as kidney function or decompensated disease, and you should not restart anything on your own. But abnormal liver enzymes alone are no longer considered a reason to stop metformin in compensated cirrhosis, and current reviews describe it as the backbone of treatment at that stage. It is a fair question to bring back to your doctor and ask directly.
If curing hepatitis C or improving my liver helps, will my diabetes go away?
Blood sugar often improves, but the diabetes usually does not disappear. Removing the cause stops further damage and can reduce insulin resistance, though scarring that has already formed continues to affect how your liver handles glucose. The realistic expectation is easier control on possibly less medicine, not a cure — and any dose reduction should be guided by readings, not assumed.
Should I ask for a CGM sensor if I have cirrhosis?
It is a reasonable thing to raise. Because HbA1c is unreliable and glucose tends to swing more in cirrhosis, continuous monitoring shows things a three-monthly lab test cannot — particularly overnight lows, which are exactly the risk that rises in liver disease. Cost is the real obstacle in Pakistan, and even a single two-week sensor used to map your pattern can be informative if ongoing use is not affordable.
Sources
- American Diabetes Association (ADA) (opens in a new tab)
- Baqai Institute of Diabetology and Endocrinology (BIDE) Pakistan (opens in a new tab)
- Pakistan Cardiac Society (PCS) / Pakistan Hypertension League (PHL) (opens in a new tab)
- American Diabetes Association (ADA) (opens in a new tab)
- American Academy of Dermatology (AAD) (opens in a new tab)
- DermNet NZ (opens in a new tab)
- Journal of Pakistan Association of Dermatologists — cutaneous manifestations of diabetes mellitus (PAF Hospital Sargodha and CMH Peshawar) (opens in a new tab)
- Kumar A, Basu R — Management of Cirrhosis in Diabetes: A Pragmatic Approach to the Patient, The Journal of Clinical Endocrinology & Metabolism (2026) (opens in a new tab)
- Hepatogenous Diabetes: A Primer — review, National Library of Medicine (PMC) (opens in a new tab)
- Advances in the treatment of hepatogenous diabetes — review, World Journal of Gastroenterology (PMC) (opens in a new tab)
- EASL Clinical Practice Guidelines on nutrition in chronic liver disease, Journal of Hepatology (PMC) (opens in a new tab)
- Hemoglobin A1c Has Suboptimal Performance to Diagnose and Monitor Diabetes Mellitus in Patients with Cirrhosis — Digestive Diseases and Sciences (opens in a new tab)
Related reading

Diabetes and Fatty Liver Disease: The Silent Connection Explained
How extra glucose is converted into stored fat inside your liver, why early fatty liver causes zero pain, and practical lifestyle habits to clear liver fat naturally.

Can Hepatitis Cause Diabetes?
Yes — but only some kinds, and mostly through two very different routes. Here is which hepatitis matters, and why the liver itself is often the real culprit.

Why Insulin and Certain Diabetes Medicines Cause Low Blood Sugar
Understand the physiological triggers behind hypoglycemia—why insulin and sulfonylureas lower sugar, and how to stay safe without fear.

Can People with Diabetes Fast During Ramadan? Understanding Your Safety Risk
A respectful, medically grounded guide to evaluating your individual risk category before Ramadan, working with your doctor, and respecting religious exemptions.